20 Prof. S. U. Pickering. Determinations to test [Dec. 11, 



0-00065 per 1 from 7'3 to 23-4 (0 = f.p.) 

 0-00153 23-4 45-4 



the average increase being large in comparison with the other sub- 

 stances examined. The valnes for the solid are somewhat irregular, 

 but, on plotting them out, they are found to be evenly distributed above 

 and below a straight line ; the value at and the rate of increase have 

 been deduced from this line, and the probable error was calculated by 

 noting the difference which would be caused by taking the values 

 above or below it. 



Comparison with the Results of other Observers. 



The results here obtained are compared in Table B with those 

 given by other observers. Except where it is stated otherwise, they 

 refer to the temperatures at which the substances melt under normal 

 conditions. The concordance exhibited by this table is in most 

 cases fairly good, though Berthelot's values for the heat of fusion of 

 sulphuric acid and its monohydrate seem wholly inexplicable. 



My value for the heat capacity of liquid calcium nitrate receives 

 confirmation from some determinations which I made some time ago 

 at 18, with various solutions of salt up to a strength of 61'4 per 

 cent. Ca(NO 3 ) 2 by an electrical method ; the results obtained formed 

 a fairly uniform curve, and, on extending this up to 69'5 per cent, 

 (the composition of the tetrahydrate), I got 0'517as the heat capacity 

 of such a solution, while the present results give 0'5185 at 42'4, or 

 0-5283 at 18. 



General Results. 



From Table A (p. 18) it will be seen that the heat capacity of solid 

 benzene is greater than that of the liquid, so that C c becomes a nega- 

 tive quantity. This is somewhat remarkable, for in other known in. 

 tttances, such as that of beeswax, where fusion is a gradual process, and 

 the heat capacity of the solid becomes abnormally great as the melting 

 point is reached, this abnormality makes itself evident in the augment- 

 ing rate at which the increase occurs, whereas with benzene no such 

 abnormal increase is noticed, although the determinations extend as 

 far as 36 below the melting point : the only sign of anything 

 unusual is that the rate of increase is considerably greater than that 

 in the other cases investigated. This observation with benzene must 

 throw some doubt on conclusions drawn from the heat capacity of any 

 solids, unles-s the determinations extend through a very long range of 

 temperature. It must be noted that the heat of fusion given here for 

 benzene will be too small if part of that heat of fusion appears as the 

 heat capacity of the solid. 



The values for the temperature of no crystallisation obtained from 

 the present rt suits aie given in Table C : the only other substances 



